 
|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
|
|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
129.44.86.229
In Reply to: RE: Best de-multiplexer to copy? posted by BofService on February 28, 2008 at 08:53:16
Most IC multiplex decoders have it all over their tubed counterparts - from a noise, distortion and especially separation perspective - plus there is only one calibration adjustment to make. HOWEVER, for best sound, I've found that you do want to follow up the chip with a tubed output stage - say an SRPP variant (need a little extra gain and lower output impedance).
Some of the later Scott and MAC tubed decoders do sound good BUT require good stereo generators to attain best performance - especially good stereo separation above 1Khz.
If you do persist, have fun finding the specs for and winding the necessary coils.
CHarles
Follow Ups:
BTW, I believe that someone over on the Yahoogroups FM tuner chatroom was building up a PC board and maybe a kit of parts for an IC decoder - don't know what chip they were using.
[there were at least two generations of IC decoders with the second haveing an order of magnitude less distortion].
The boards I like to use came from VSM audio on Long Island - he originally used the LM1700 (I believe) first generation chip and then changed to a second generation Toko chip. Uses a voltage doubler to derive 12 VDC power from a 6.3VAC (tube filament) source so it can (fairly easily) mount in a mono (tubed) tuner.
Cheers
The board you're referring to is built around the LM4500 chip, which is a later generation MPX chip with low distortion and anti-birdie circuitry.
from this site
http://rbsfm.ej.am/joomla/index.php?option=com_content&task=view&id=53&Itemid=30
LM4500 - FM Stereo Demodulator
Written by BuSan
Tuesday, 21 November 2006
Description
"The LM4500 is an improved stereo demodulator IC offering very low audio distortion. A new demodulator technique minimizes adjacent station interference caused by subcarrier harmonics and prevents lock-up problems from pilot carrier frequency harmonics. The IC features a blend circuit which optimizes the signal-to-noise ratio under weak signal conditions by gradually combining left and right channel information."
I measured .035% distortion driven by a KT-8300 at full modulation. I designed the LM4500 board mostly for my own use, and then to supply friends who wanted it. The site for info is www.FMMPX.com
Haven't been following this on the Yahoogroups chatroom.
Are there / can you point me to recommended mpx filter schematics?
Charles
Hi Charles;
The output filter is still evolving, I'm doing listening tests now, almost done. There has been a lot of discussion about the philosophy of it. So I released the LM4500 board per the data sheet schematic with de-emphasis implemented onboard. Some viewpoints are:
1) you don't need any filtering, but should do the de-emphasis passively. This person posted a design in the Yahoo FMTuners group files using passive de-emphasis followed by FET buffers.
2) Don't care about passive de-emphasis, but you should use a sharp brick wall filter after 15kHz. There were two diverging opinions about implementation, one used many opamps, the other used passive L,C components. There were designs posted of these as well.
3) You should do passive de-emphasis, and use notch filters at 19 and 38kHz to kill most the highest levels of ultrasonic garbage. This sidesteps the phase shift inherent in brickwall filters.
This last approach 3), from my friend Ray, is the board I have been developing. It is fairly simple and the schematic is shown in the picture on the bottom of page 1 of the fmmpx.com site.
It's design is as follows - input, series 3.3k resistor, followed in series by 15mHy in Parallel with .0047uF (19kHz notch). This is followed by .0012uF and 15mHy in series to ground (38kHz notch). Then a 10k resistor to gnd, followed by a series 10k resistor, and then .0068uF cap to gnd. This last stage is the passive 75uS de-emphasis. This is then followed by an output stage with gain, that has an input impedance of 100k - it can be tube or ss, but the board I will make available is simply 2 single op-amps (L,R) with some gain. I always felt if you must us op-amps, at least use singles with separate supplies and de-coupling.
I went into this effort thinking of it as a DIY club where people make their own choices, versus a finished product where I make them. The funny part is almost everyone wants the latter. I'm about to lock in the design and have boards made for this output filter, but want some people besides me to listen to it first.
Bob
OK Bob,
I intuitively agree with your approach three, however my own listening tests haven't been able to discern any audible benefit from filtering. Sure does clean up test signals on a 'scope though.
Charles
Charles;
Again a somewhat philosophical discussion. Many notable tuners, including some McItosh solid state models, shipped with no output filters.
The filters are really only *needed* if you are recording the FM program with an analog tape deck. Or if you *feel* you need it after looking at the scope trace of the output, (or your wife, kids or dog can hear 19kHz) :-)
I would say build it without first, then add the filter board later to see if you can hear the difference.
The minimalist LM4500 MPX board with de-emphasis implemented, and no additional buffer or circuits, sounds very good indeed. Done that way it's a pretty cheap and easy to build MPX solution, too. Beyond the board and LM4500 chip, the parts needed from Digi-key cost about $15.
Bob
I agree, finding the transformers to build your own would be a big problem.
I was thinking about this a long time ago, and bought a few Fisher tube MPX sections from a guy that had gutted Fisher 500B receivers. This was back when nobody cared too much for them, and the value of one with bad tubes was pretty low. The MPX section was made as a separate unit, which then bolted onto the receiver and wired up. Rumor was they were made in Germany, but not sure if it's true. The later MPX circuit that Fisher used in their receivers and tuners looks like a copy of the Scott design to me. I may have a spare one of these laying around still.
Hi, BFitz:
I don't think that it was a rumor that the drop-in mulitiplex portion of the old Fisher receivers was manufactured in Germany.
Fisher used a lot of German-made parts in his receivers: miniature tubes, capacitors, selenium rectifiers, etc. The knobs were made in W. Germany, but the meters, oddly enough, were usually made in Japan.
Sort of a "United Nations" kind of an effort, eh?
Whatever worked and was more or less cheap and reliable, was what went in there.
BTW, I do not think that the engineers at Fisher Radio Corporation were copying any Scott designs way back then. That would have been just a bit too close for comfort, wouldn't you think?
Richard Links
Berkeley, CA
Richard;
Fisher changed their MPX design mid way through the MPX100 production. They switched from a less effective circuit to the MPX topology pioneered and first used by Scott. The Fisher design is not an exact copy of the Scott, but is quite similar overall. They both use the same type of 8 diode switching circuit that eventually became the standard for all the better tube MPX decoder designs.
Thanks, BFitz:
Perhaps in that era, this was a bit more of "friendly borrowing" than stealing.
Sort of like the similarity in ciruit designs amongs the following preamps:
McIntosh C-11 C-22
Marantz Model 7
Richard Links
Berkeley, CA
| FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
||||||||||||||
This post is made possible by the generous support of people like you and our sponsors:
Top of Page ]
[ Contact Us ]
[ Support/Wish List ]
[ Copyright © 2025, Audio Asylum, all rights reserved ]
